Project summary: Retroviruses such as HIV are detrimental to mankind, killing and infecting millions globally. An essential protein for replication and maturation of retroviruses are aspartate proteases (Pr). These proteases are symmetric dimers, where each monomer contributes a beta hairpin flap that acts as a gatekeeper for substrates to the active sites. Currently there are 9 protease inhibitors approved by the FDA to treat HIV, but none exist to treat non HIV retroviruses. All 9 approved protease inhibitors are active site inhibitors, developed by optimizing non- bonded interactions with amino acids in the active site. Unfortunately drug resistant mutants exist for all 9 active site inhibitors. To combat drug resistance, researchers have focused on developing new inhibition strategy that target non-active site residues in retroviral proteases. Using MD simulations and EPR experiments, we showed that the mechanism of substrate binding in HIV-1 Pr requires flap opening and partial dissociation of the dimer interface, leading to a transient, solvent exposed pocket. We refer to this state as a partially dissociated dimer, because the interleaved beta sheet termini accounting for 75 % of dimer stabilizing energy is still intact. Our goal is to screen for small molecules that can perturb flap dynamics and active site geometry by binding into the solvent exposed pocket in the dimer interface of retroviral proteases. Our central hypothesis is that a small molecule can bind in the pocket and allosterically prevent flap closure. Our rationale is once the flaps remain open, the active site geometry will be incorrect for binding of the natural substrate, resulting in decreased catalytic efficiency of the protease. To test our hypothesis, we will use MD simulations to characterize the dimer interface pocket of multiple sequences of HIV-1 proteases. Although the dimer interface is conserved we will like to investigate the effects of current clinical MDR mutations on pocket formation and geometry (Aim 1). Next, we will investigate whether our allosteric mechanism of inhibition is also relevant to non HIV-1 proteases, since using active site inhibitors can be challenging due to differences in substrate recognition across retroviral proteases. We will use HTLV-1 Pr and rous sarcoma virus Pr to characterize in detail their opening mechanism/partial dissociation, which is currently unknown (AIM 2). Finally, we will computationally screen allosteric inhibitors that can stabilize the inactive open flap conformations by binding to the dimer interface pocket in retroviral proteases (Aim 3). The probability of developing drug resistance to our inhibitors will be smaller than for current inhibitors since this region is highly conserved to preserve functionally important dynamics (substrate entry). Conclusion of our work will provide: 1) a new inhibition strategy, details on a druggable pocket, and small molecule candidates to combat MDR in retroviral proteases, and 2) atomistic detail on the opening mechanism of non HIV-1 proteases to assess the feasibility of targeting them with a similar allosteric inhibition strategy to what we will document for HIV-1 protease.